Clay

Run data quality checks on PMU (Phasor Measurement Unit) data. Use when the user asks to validate, check, or audit PMU measurements including frequency, volt...

Security Scan

The skill's code and runtime instructions are coherent with its stated purpose (local PMU CSV data quality checks), request no credentials, and don't appear to exfiltrate data or perform unrelated actions.

Purpose & Capability

Name/description match the included files: SKILL.md, a sample CSV, a limits config, and a Python script that implements frequency, voltage, phasor angle, missing-data, and timestamp-gap checks. All requested actions (reading a CSV, applying configurable limits, producing reports) are appropriate for a PMU data quality tool.

Instruction Scope

Runtime instructions restrict the skill to asking for a PMU CSV, checking columns, running the provided Python script, and reporting local outputs. The SKILL.md does not instruct the agent to read unrelated system files, environment variables, or send data to external endpoints.

Install Mechanism

This is an instruction-only skill with an included Python script (no install spec). The script imports pandas and numpy but the skill does not declare these dependencies or require 'python' as a runtime binary; users should ensure a compatible Python environment with required packages is available before running.

Credentials

The skill requests no environment variables, credentials, or config paths. All file access is limited to user-supplied CSVs, the included templates, and locally produced reports, which is proportionate to the stated purpose.

Persistence & Privilege

always is false and the skill does not request persistent or elevated privileges. It does write output files (flagged CSV and optional HTML) alongside input files, which is expected for a reporting tool.

Assessment

This skill appears to be a straightforward local PMU CSV quality checker. Before installing/running: 1) Review the full script if you need to be certain it won't access other files or networks (the visible code imports pandas/numpy and writes local reports). 2) Run it in an environment with Python and the required libraries (pandas, numpy) installed or in a sandbox if the data is sensitive. 3) Be aware it reads whatever CSV path you provide and writes flagged output next to it — avoid pointing it at directories containing unrelated sensitive files. 4) If you plan to use it in production, confirm the limits_config.json matches your system and review any further code not shown to ensure no unexpected network or subprocess calls.

PMU Data Quality Checker

Performs automated data quality checks on PMU (Phasor Measurement Unit) CSV data files against configurable security limits defined in IEEE/NERC standards.

What This Skill Checks

1. Frequency Data — Checks if frequency measurements stay within the nominal range (default: 59.95–60.05 Hz for 60 Hz systems)
2. Voltage Magnitude Data — Checks if voltage magnitudes stay within acceptable per-unit limits (default: 0.95–1.05 pu)
3. Phasor Angle Data — Checks if phasor angles stay within expected bounds (default: -180° to +180°, with rate-of-change check)
4. Missing / NaN Data — Flags rows with missing or null values
5. Timestamp Continuity — Detects gaps in the reporting rate (e.g., expected 30 samples/sec or 60 samples/sec)

How to Use

Quick check on a CSV file:

python <skill_base_path>/scripts/pmu_quality_check.py path/to/data.csv

With custom limits config:

python <skill_base_path>/scripts/pmu_quality_check.py path/to/data.csv --config <skill_base_path>/templates/limits_config.json

On the template sample data (for testing):

python <skill_base_path>/scripts/pmu_quality_check.py <skill_base_path>/templates/sample_pmu_data.csv

Expected CSV Format

The input CSV should have columns similar to:

timestamp	frequency	voltage_mag	voltage_angle	current_mag	current_angle

• timestamp — ISO 8601 or Unix epoch
• frequency — in Hz
• voltage_mag — in per-unit (pu) or kV (specify in config)
• voltage_angle / current_angle — in degrees
• current_mag — in per-unit (pu) or Amps

Column names are configurable via the limits config JSON. If the user's CSV uses different column names (like those from openHistorian or FNET exports), update the column_mapping section in the config.

Output

The script produces:

• A summary report printed to stdout
• A flagged rows CSV saved alongside the input (e.g., data_flagged.csv)
• An optional HTML report with charts if --html flag is passed

Workflow

1. Ask the user for their PMU data file (CSV)
2. Check if the column names match the expected format; if not, ask or auto-detect
3. Run the quality check script
4. Present the summary and ask if the user wants to adjust limits or dig deeper into flagged data